Topics in statistics communication: interpretation of probability phrases and factchecking misleading bar charts
Sanne Willems, Leiden University
Verbal probability phrases, as “unlikely”, “usually” and “maybe”, are often used in science communication to express estimated risks in words instead of numbers. In our study we looked at how laypeople and statisticians interpret Dutch probability phrases that are regularly used in news articles. In the first half of my talk, I will explain more about the set-up of this study and present the results. I will show the variability in the numerical interpretation of these phrases, discuss (a)symmetry in the interpretations and compare the results of statisticians and non-statisticians.
In the second part, I will talk about a second study in statistics communication, namely on methods to combat misleading graphs. Graphs are useful to communicate concisely about complex issues to a broad public. However, many misleading graphs pop-up on social media feeds and hasty readers often draw the wrong conclusions if graphs are misleading due to violated design conventions. For example, omitting the baseline of the vertical axis of a bar chart is a common trick to exaggerate differences between groups. We investigated and compared the effectiveness of four correction methods as debunking strategies for these misleading bar charts, each focusing on different phases of graph reading. I will show the set-up of the study and discuss the findings.
*please check your spam folder for the confirmation message after subscribing, and contact us if you haven't received a message within a day.
- Begindatum en -tijd
- Einddatum en -tijd
- Online in Microsoft Teams